A Visual Inspection Tool for Evaluation of ASR Model Using PyKaldi and PyCHAIN

Jarin, Asril; Santosa, Agung; Aini, Lyla Ruslana; Uliniansyah, Mohammad Teduh; Nurfadhilah, Elvira; Isnaeni, Dian; Pebiana, Siska; Desiani,; Gunarso,

doi:10.1109/icitacee55701.2022.9924132

Cited by 1 publication

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The ASR model is trained using Kaldi and PyChain, and its testing and evaluation are carried out iteratively by measuring the WER value. During the evaluation process, the alignment of transcripts and speech phonemes recognized by ASR is visually inspected using an evaluation tool [19]. The evaluation results provide feedback to improve  ISSN: 2252-8938 the data preparation process, optimize the language model specification and lexicon, and refine data augmentation techniques.…”

Section: Asr Development For Spwpm Servicementioning

confidence: 99%

“…To analyze and understand the reasons behind ASR inaccuracies, researchers need a tool that enables visual examination and monitoring of the data generated during the ASR recognition process, including transcripts and speech alignment. A tool developed by Jarin et al [19] serves this purpose, providing researchers with assistance. This tool utilizes the PyKaldi [33] and PyChain modules to display speech waveforms, and MFCC features on an HTML5 canvas.…”

Section: Evaluating Asr With a Kaldi-based Asr Evaluation Toolmentioning

confidence: 99%

See 1 more Smart Citation

Automatic speech recognition for Indonesian medical dictation in cloud environment

Jarin,

Santosa,

Uliniansyah

et al. 2024

IJ-AI

View full text Add to dashboard Cite

<span lang="EN-US">This paper introduces SPWPM, an automatic speech recognition (ASR) system designed specifically for Indonesian medical dictation. The main objective of SPWPM is to assist medical professionals in producing medical reports and diagnosing patients. Deployed within a cloud computing service architecture, SPWPM strives to achieve a minimum speech recognition accuracy of 95%. The ASR model of SPWPM is developed using Kaldi and PyChain technologies—creating a comprehensive training dataset involving collaboration with PT Dua-Empat-Tujuh and Harapan Kita Hospital. Several optimization techniques were applied, including language modeling with smoothing, lexicon generation using the Grapheme-to-Phoneme Converter, and data augmentation. The readiness of this technology to assist hospital users was assessed through two evaluations: the SPWPM architecture test and the SPWPM speech recognition test. The results demonstrate the system's preparedness in accurately transcribing medical dictation, showcasing its potential to enhance medical reporting for healthcare professionals in hospital environments.</span>

show abstract

Section: Asr Development For Spwpm Servicementioning

confidence: 99%

Section: Evaluating Asr With a Kaldi-based Asr Evaluation Toolmentioning

confidence: 99%